1 This file records noteworthy changes.
4 * interface cleanups and bugfixes.
7 * series expansion now much more consistent for small order expansion.
8 * lsolve() accepts algorithmic hint as parameter.
11 * Epsilon tensor is more functional.
12 * simplify_indexed() is better at detecting expressions that vanish for
14 * Several little bugfixes and consistency enhancements.
17 * Ctors of class numeric are not explicit any more. All built-in callers for
18 pseudofunctions are now templated and default to ex arguments which relaxes
19 the need for explicit ctors.
20 * New functions/methods:
22 - remove_first(), remove_last(), sort() and unique() for lists
25 * Instead of just totally symmetric or antisymmetric, complex symmetries
26 can now be defined for indexed objects. Symmetries are described by a
27 tree of "symmetry" objects that is constructed with the sy_none(),
28 sy_symm(), sy_anti() and sy_cycl() functions. The symmetry of a function
29 with respect to its arguments can also be defined (this is currently
30 only used for the Beta function).
31 * Generalized map() to take a function object instead of a function pointer.
32 This allows passing an arbitrary number of additional state to the
33 function being called.
34 * color_trace(), dirac_trace(), diff(), expand(), evalf() and normal() work
35 better with container classes, e.g. using color_trace() on a relation will
36 take the trace on both sides, using diff() on a matrix differentiates every
38 * diff() works properly with non-commutative products and indexed objects.
39 * New option flag "expand_function_args" for expand().
40 * Supplement some (now deprecated) macros by inlined template functions:
41 - is_of_type(foo, type) -> is_a<type>(foo)
42 - is_ex_of_type(foo, type) -> is_a<type>(foo)
43 - is_exaclty_of_type(foo, type) -> is_exaclty_a<type>(foo)
44 - is_ex_exaclty_of_type(foo, type) -> is_exaclty_a<type>(foo)
45 - ex_to_foobar(baz) -> ex_to<foobar>(baz)
46 * rem(c, p[x], x) (c: numeric, p[x]: polynomial) erroneously returned p[x]
48 * Small bugfixes in pattern matching.
49 * Updated libtool to version 1.4.
52 * In the output and in ginsh, lists are now delimited by { } braces, and
53 matrices are delimited by single [ ] brackets.
54 * simplify_indexed() renames dummy indices so, e.g., "a.i*a.i+a.j*a.j" gets
55 simplified to "2*a.i*a.i" (or "2*a.j*a.j").
56 * New functions/methods:
57 - canonicalize_clifford() (helpful when comparing expressions containing
59 - symmetrize() and antisymmetrize()
60 - numer_denom() (return numerator and denominator in one call)
61 - map() (apply function to subexpressions)
62 - evalm() (evaluate sums, products and integer powers of matrices)
63 * Added a new function match() for performing pattern matching. subs() and
64 has() also accept patterns as arguments. A pattern can be any expression,
65 optionally containing wildcard objects. These are constructed with the
66 call "wild(<unsigned>)" and are denoted as "$0", "$1" etc. in the output
68 * Positive integer powers of non-commutative expressions (except matrices)
69 are automatically expanded.
70 * Removed cint subdirectory, ginaccint is a separate package now due to
71 packaging considerations.
72 * Several little bugfixes.
75 * color and clifford classes are functional and documented.
76 * New "spinidx" class for dotted/undotted indices.
77 * Predefined spinor metric tensor (created by spinor_metric()).
78 * Symbols can have a LaTeX name, e.g. symbol s("s", "\\sigma");
79 * LaTeX output of indexed objects is much nicer.
80 * Fixed some build problems (with recent libreadline).
81 * Semantics of arithmetic operators now follows the C++ rules more strictly.
84 * degree(), ldegree(), coeff(), lcoeff(), tcoeff() and collect() work with
85 non-symbols as the second argument in ginsh.
86 * the argument to collect() can be a list of objects in which case the
87 result is either a recursively collected polynomial, or a polynomial in
88 a distributed form with terms like coeff*x1^e1*...*xn^en, as specified by
89 the second argument to collect().
90 * Several bugfixes (including a nasty memory leak in .normal()).
91 * class matrix: solve() doesn't call algorithms redundantly any more and
92 inverse() falls back to solve() which works in more general cases.
95 * degree(), ldegree(), coeff(), lcoeff(), tcoeff() and collect() can now
96 be used with constants, functions and indexed expressions as well, so you
97 can use it to collect by powers of Pi or sin(x), or to find the coefficient
100 - it only works with symbols, constants, functions and indexed expressions,
101 trying to find the coefficient of, e.g., "x^2" or "x+y" won't work;
102 - it does not know about dummy index summations; the coefficient of
103 gamma~0 in p.mu*gamma~mu should be p.0 but is returned as 0;
104 - using coeff(), tcoeff(), lcoeff() or collect() on elements of
105 noncommutative products might return wrong or surprising results.
106 * subs() no longer only substitutes symbols and indices but performs a more
107 general "syntactic substitution", i.e. it substitutes whole objects in sub-
108 expressions. You can subs((a+b)^2,a+b==3) and get 9, but subs(a+b+c,a+b==3)
111 - substituting numerics (subs(expr, 2==4)) will not replace then in all
112 occurences; in general, you shouldn't substitute numerics, though.
113 * Added preliminary (re)implementations of color and clifford classes.
114 * simplify_indexed(): contraction of symmetric and antisymmetric tensors
116 * Replaced the various print*() member functions by a single print() that
117 takes a print_context object that determines the output formatting. This
118 should make it easier to add more output types, such as LaTeX output,
119 which is based on work by Stefan Weinzierl.
120 * Added functions to retrieve the properties stored in archive objects
121 outside of unarchive() (for printing or debugging purposes).
122 * Some bugfixes (indexed objects, archive writing).
123 * .collect() on non-polynomials is now algebraically correct.
125 0.8.0 (24 March 2001)
126 * Complete revamp of indexed objects. Instead of multiple classes for
127 indexed things and their indices there is now only one "indexed" class
128 and two types of indices: "idx" for simple indices and "varidx" for
129 indices with variance. There are predefined delta, epsilon and metric
130 tensors, and a function simplify_indexed() that performs canonicalization
131 and dummy index summations. Matrix objects can be indexed for doing simple
133 * Added an option "expand_indexed" to expand() to perform expansion of
134 indexed objects like (a+b).i -> a.i + b.i
135 * Renamed get_indices() to get_free_indices(), which no longer returns
136 dummy indices and checks the consistency of indices in sums.
137 * sqrfree() factorization fixed and improved syntactically.
138 * subs() works on matrices.
139 * Matrices can be constructed from flat list of elements; diagonal matrices
140 can be constructed from list of diagonal elements with diag_matrix().
141 * Fixed memory leak in expand().
142 * Operator% for objects of class ncmul has gone. Use operator* now for that
143 case too, which is much more natural.
145 0.7.3 (28 February 2001)
146 * Several bugfixes and minor performance tunings.
147 * Added a section to the tutorial about adding new algebraic classes to GiNaC.
148 * Closed many in-source documentation gaps.
150 0.7.2 (17 February 2001)
151 * Several bugfixes in power series expansion, one of them critical.
153 0.7.1 (7 February 2001)
154 * Fix problems with Cint that were caused by CLN's overloaded operator new.
155 * Fix compilation errors with GCC3.
156 * normal() handles large sums of fractions better and normalizes the exponent
157 of power expressions.
158 * expand() always expands the exponent and transforms x^(a+b) -> x^a*x^b.
159 * Some bugfixes of series expansion around branch cuts of special functions.
161 0.7.0 (15 December 2000)
162 * Requires CLN 1.1 now. Class numeric doesn't use an indirect pointer to the
163 actual representation any more. This is a speedup.
164 * mul::expand() was reengineered to not allocate excess temporary memory.
165 * Non-integer powers of a symbol are treated as constants by (l)degree() and
166 coeff(). Using these functions on an expression containing such powers used
167 to fail with an internal error message. The side-effect is that collect()
168 can be used on expressions which are not polynomials.
169 * Added a man page for the ginac-config script.
170 * Ctor of numeric from char* honors Digits.
172 0.6.4 (10 August 2000)
173 * Complete revamp of methods in class matrix. Some redundant (and poor)
174 implementations of elimination schemes were thrown out. The code is now
175 highly orthogonal, more flexible and much more efficient.
176 * Some long standing and quite nasty bugs were discovered and fixed in the
177 following functions: add::normal(), heur_gcd(), sr_gcd() and Order_eval().
180 * Derivatives are now assembled in a slightly different manner (i.e. they
181 might 'look' different on first sight). Under certain circumstances this
182 can result in a dramatic speedup because it gives hashing a better chance,
183 especially when computing higher derivatives.
184 * Some series expansions of built-in functions have been reengineered.
185 * The algorithm for computing determinants can be chosen by the user. See
186 ginac/flags.h and ginac/matrix.h.
187 * The Dilogarithm (Li2) now has floating point evaluation, derivative and a
188 proper series expansion.
189 * Namespace 'std' cleanly disentangled, as demanded by ISO/EIC 14882-1998(E).
190 * Some minor bugfixes, one major lsolve()-bugfix and documentation updates.
193 * ginaccint.bin is now launched by a binary program instead of by a scripts.
194 This allows us to write #!-scripts. A small test suite for GiNaC-cint was
196 * Several minor bugfixes.
199 * Cleanup in the interface to Cint. The required version is now Cint 5.14.38.
200 * Several bugfixes in target install.
203 * IMPORTANT: Several interface changes make programs written with GiNaC
204 much clearer but break compatibility with older versions:
205 - f(x).series(x,p[,o]) -> f(x).series(x==p,o)
206 - series(f(x),x,p[,o]) -> series(f(x),x==p,o)
207 - gamma() -> tgamma() (The true Gamma function, there is now also
208 log(tgamma()), called lgamma(), in accord with ISO/IEC 9899:1999.)
209 - EulerGamma -> Euler
210 * #include'ing ginac.h defines the preprocessor symbols GINACLIB_MAJOR_VERSION,
211 GINACLIB_MINOR_VERSION, and GINACLIB_MICRO_VERSION with the respective GiNaC
212 library version numbers.
213 * Expressions can be constructed from strings like this:
214 ex e("2*x+y", lst(x, y));
215 * ex::to_rational() provides a way to extend the domain of functions like
216 gcd() and divide() that only work on polynomials or rational functions (the
217 good old ex::subs() method reverses this process)
218 * Calling diff() on a function that has no derivative defined returns the
219 inert derivative function "Derivative".
220 * Several new timings in the check target. Some of them may be rather rude
221 at your machine, feel free to interrupt them.
223 0.5.4 (15 March 2000)
224 * Some algorithms in class matrix (notably determinant) were replaced by
225 less brain-dead ones and should now have much better performance.
226 * Checks were completely reorganized and split up into three parts:
227 a) exams (small regression tests with predefined input)
228 b) checks (lenghty coherence checks with random input)
229 c) timings (for coherence and crude benchmarking)
230 * Behaviour of .evalf() was changed: it doesn't .evalf() any exponents.
231 * Expanded expressions now remember they are expanded to prevent
232 superfluous expansions.
233 * Small bugfixes and improvements in the series expansion.
235 0.5.3 (23 February 2000)
236 * A more flexible scheme for registering functions was implemented,
237 allowing for remembering, too.
240 0.5.2 (16 February 2000)
241 * Mainly fixes a bug in the packaging of release 0.5.1.
243 0.5.1 (14 February 2000)
244 * Fixes a small number of bugs.
246 0.5.0 (7 February 2000)
247 * Expressions can be written ("archived") to files and read therefrom.
248 * Addition of GiNaC-cint, which lets you write complete programs in
249 an interactive shell-like manner in your favoured programming
252 0.4.1 (13 December 1999)
253 * Series Expansion of Gamma function and some other trigonometric
254 functions at their poles works now.
255 * Many more evaluations of special functions at points where
257 * info_flags::rational doesn't return true for complex extensions
258 any more---use info_flags::crational for the old behaviour.
259 info_flags::integer and -::cinteger work similarly, the same
260 holds for types like info_flags::rational_polynomial.
262 0.4.0 (26 November 1999)
263 * First public release.